現在網路直播與影像串流已大量的融入人們的生活中，而直播影像大多都是透過CDN(內容分發網路)來搭建，但隨著使用者的激增與服務品質保證的考量，若只是透過增加硬體設備來減緩網路頻寬壅塞，容易造成營運成本及維護上的困難。vCDN則是上述問題的解決方式之一，vCDN(虛擬化CDN) 利用虛擬化技術將實體主機資源虛擬化，根據客戶對服務的要求，為客戶分配相應的資源和網路，以提供靈活的內容分發服務。此外，透過引入SDN技術可以因應用戶差異化及動態變化的業務需求，進而提供快速精準的服務。
因此，為了讓目前的多媒體串流環境更加地友善，讓使用者能夠更加快速地取得多媒體資源，本論文提出了「基於軟體定義網路(SDN)與 vCDN 整合之多媒體串流傳輸最佳化研究」，主要目的是透過引入網路功能虛擬化技術(NFV)，為多個用戶提供內容分發服務(vCDN)，達成資源共享的多媒體串流服務，並利用軟體定義網路(SDN)以及多重路徑演算法，在確保串流傳輸效率與服務品質(QoS)的前提下，透過適當 vCDN之部屬，在兼顧系統成本的同時達成用戶等級要求(SLA)，完成以 SDN 與 vCDN 整合網路架構為主的高效率多媒體串流服務。
實驗結果顯示，此研究改善了傳統網路固有的問題，不僅減少了設備成本，還能增加網路使用效率，使影像串流更順暢。本論文對vCDN研究效益，不僅只帶來 CDN的虛擬化，還有網路功能虛擬化的應用探討，除了可以減少許多不必要的資源浪費，對於我們生活上的便利也有相當大的好處。

Abstract:
Nowadays, live webcast and video streaming have been widely used in people's daily lives, and most live or video streaming are delivered through CDN (content delivery network). With the increasing of users and the consideration of service quality assurance, deploy more hardware and equipment is a way to fulfill the demand and to ease the network congestion. However, the operating costs and maintenance difficulties are also increasing while adding more hardware to the infrastructure. Therefore, vCDN is one of the solutions to the above problem. vCDN (virtualized CDN) uses virtualization technology and a unified control platform to isolate the physical resources such as service host and network bandwidth to provide content distribution services according to the users’ requirements for services, and so as to the resources with flexibility and agility. In addition, with the introduction of SDN technology, it can provide fast and accurate services according to the differentiated and dynamic needs of users. 
Therefore, in order to make the current multimedia streaming environment friendlier and allow users to obtain multimedia resources more quickly, this paper proposes "optimization of multimedia streaming transmission based on the integration of software-defined networking (SDN) and vCDN". First, the main purpose of this thesis is to provide content delivery services (vCDN) for multiple users through the introduction of network function virtualization technology (NFV), to achieve resource sharing. Second, use the technique of software-defined networking (SDN) and appropriate routing algorithms to ensure the premise of quality of service (QoS). Third, deploy vCDN appropriately with the consideration of the system cost and user-level requirements (SLA) to integrate SDN architecture with vCDN to provide cost-aware and high-efficiency multimedia streaming services.
The experimental results show that this research improves the inherent problems of the traditional network, not only reduces the cost of equipment but also increases the efficiency of the network, making the streaming smoother. This research not only bring about the virtualization of CDN based on SDN/NFV and the application of virtualization of network functions, which reduce the unnecessary waste of resources but also has considerable benefits for the convenience of our life.

致謝	I
中文摘要	Ⅱ
英文摘要	Ⅲ
目錄	IV
圖目錄	VII
表目錄	IX
第一章	緒論 1
1.1	前言 1
1.2	動機與目的 2
1.3	論文章節架構 4
第二章	相關研究與背景資料 5
2.1	Content Delivery Network (CDN) 5
2.1.1	Network Function Virtualization(NFV) 6
2.1.2	CDN virtualization 10
2.2	SDN routing mechanism 12
2.2.1	Dijkstra's algorithm 12
2.2.2	Probability based Selection 13
2.3	SDN/NFV and vCDN 14
2.4	Service Level Agreement (SLA) 16
第三章	vCDN之多媒體串流可調式路由機制 19
3.1 	系統架構 20
3.2 	可調式路由機制Adjustable Routing Mechanism 23
3.2.1 	利潤判斷模組 23
3.2.2 	路由機制模組 26
3.2.2.1	Probability Based Selection(PBS) 27
3.2.2.2	Bandwidth Based Selection(BBS) 28
3.2.3 	策略分配與利潤計算模組 29
3.2.3.1 開啟vCDN的成本計算 31
3.2.3.2 重整路由的成本計算 31
3.2.3.3 關閉vCDN的成本計算 32
3.2.3.4 策略分配	33
3.3	 整體流程整合 35
第四章	實驗結果與效能分析 37
4.1	實驗環境架構 37
4.1.1	硬體規格	38
4.1.2	系統參數	39
4.1.3	實驗環境與網路拓撲 40
4.2	數據與分析 42
4.3	實驗成果展示 57
4.4	主要貢獻	59
第五章	結論與未來展望 60
參考文獻	61

圖目錄
圖2.1	CDN示意圖 5
圖2.2	透過NFV虛擬化的硬體和軟體	6
圖2.3	NFV 架構框架 7
圖2.4	找最短路徑及其最短路徑樹(以A點為來源) 13
圖3.1	系統架構圖 20
圖3.2	利潤判斷模組流程圖 25
圖3.3	路由機制模組的流程圖 28
圖3.4	策略分配與利潤計算模組流程圖 33
圖3.5	關閉vCDN策略流程圖 34
圖3.6	系統流程圖 36
圖4.1	實驗架構圖 37
圖4.2	網路拓撲圖 40
圖4.3	PBS同時提供9條服務-吞吐量	43
圖4.4	PBS同時提供9條服務-遺失率	43
圖4.5	ARM同時提供9條服務-吞吐量	45
圖4.6	ARM同時提供9條服務-遺失率	45
圖4.7	ARM與PBS方法平均遺失率比較 46
圖4.8	ARM與PBS方法於用戶數上的比較結果 47
圖4.9	客戶類型人數分布 49
圖4.10	系統服務客戶類型平均接受率(Parameter A) 51
圖4.11	系統服務客戶類型平均接受率(Parameter B) 51
圖4.12	系統整體平均拒絕率(Parameter A) 52
圖4.13	系統整體平均拒絕率(Parameter B) 52
圖4.14	拒絕率隨著閾值X變化之變動情形(Parameter A) 53
圖4.15	拒絕率隨著閾值X變化之變動情形(Parameter B) 54
圖4.16	系統資源利潤PA隨著不同類型客戶比例變動情形	55
圖4.17	Parameter A、Parameter B縮短與理想利潤值差距百分比 56
圖4.18	PBS同時9個用戶 58
圖4.19	ARM同時9個用戶 58

表目錄
表3.1	路徑負載參數 26
表4.1	硬體規格表 38
表4.2	系統參數	39
表4.3	客戶循環出現示意表 41
表4.4	客戶使用頻寬 41
表4.5	PBS同時提供9條服務 42
表4.6	ARM同時提供9條服務 44
表4.7	驗證負載平衡數據統計 46
表4.8	ARM與PBS方法於利潤上的比較結果 47
表4.9	系統參數種類 48
表4.10	客戶類型分佈比例	49

[1]	Bradley Mitchel.(2018).Content Delivery and Distribution Networks (CDN).
Retrieved from:　https://www.lifewire.com/introduction-to-cdn-817799 (Dec. 10,　2018), from Lifewire.
[2]	Cisco , (2020)Cisco Annual Internet Report (2018–2023)[White Paper], March 9, 2020.
[3]	WEI-CHUN CHENG.  “Quality aware routing for video streaming over SDN” .Tamkang University, 2015, Master’s Thesis.
[4]	KUI-RONG CHEN.”SELF-ADAPTIVE NETWORK QOS ROUTING MECHANISM”. Tamkang University, 2018, Master’s Thesis.
[5]	ADMIN BESTARIWEB.(2016).Cara membuat CDN Sendiri dengan BIND dan GeoIP . 
Retrieved from : https://blog.bestariwebhost.com/cara-membuat-cdn-
sendiri-dengan-bind-dan-geoip/, “Mengenal Apa itu CDN dan Bagaimana cara membuat CDN Server sendiri ”.
[6]	F5.(2013).Network Functions Virtualization - Everything Old Is New Again. Retrieved from : https://www.f5.com/services/resources/
white-papers/network-functions-virtualization-everything-old-is-new-again (Dec. 11,2018).
[7]	Margaret Chiosi, Steve Wright.(2013,Oct.). “Network Functions Virtualisation (NFV). SDN and OpenFlow World Congress”. Frankfurt-Germany .
[8]	N. Bouten et al., “Towards NFV-based multimedia delivery, ” 2015 IFIP/IEEE International Symposium on Integrated Network Management (IM), Ottawa, ON, 2015, pp. 738-741.
[9]	E. Dijkstra, “A note on two problems in connexion with graphs,” Numerische mathematik, vol. 1, no.1, pp. 269-271, 1959.
[10]	S. Van Rossem et al., “Deploying elastic routing capability in an SDN/NFV-enabled environment, ” 2015 IEEE Conference on Network Function Virtualization and Software Defined Network (NFV-SDN), San Francisco, CA, 2015, pp. 22-24.
[11]	ONF Solution Brief, “OpenFlow-enabled SDN and Network Functions Virtualization”, Open Networking Foundation, Feb. 17,2014.
[12]	L. Peterson et al., “Central office re-architected as a data center, ” in IEEE Communications Magazine, vol. 54, no. 10, pp. 96-101, October 2016.
[13]	B. Sotomayor, R. S. Montero, I. M. Llorente and I. Foster, “Virtual Infrastructure Management in Private and Hybrid Clouds, ” in IEEE Internet Computing, vol. 13, no. 5, pp. 14-22, Sept.-Oct. 2009.
[14]	M. Macías, J. O. Fitó and J. Guitart, “Rule-based SLA management for revenue maximisation in Cloud Computing Markets, ” 2010 International Conference on Network and Service Management, Niagara Falls, ON, 2010, pp. 354-357.
[15]	C. Xiao, G. Chang and B. Bing, “OPN09-05: An SLA-Aware Transport Protocol for High Throughput Wide Area Ethernet Services, ” IEEE Globecom 2006, San Francisco, CA, 2006, pp. 1-5.
[16]	L. Wu, S. K. Garg, S. Versteeg and R. Buyya, “SLA-Based Resource Provisioning for Hosted Software-as-a-Service Applications in Cloud Computing Environments,” in IEEE Transactions on Services Computing, vol. 7, no. 3, pp. 465-485, July-Sept. 2014.
[17]	D. Mensce and V. Almeida, Capacity Planning for Web Performance: Metrics Models and Methods, USA, NJ, Upper Sadale River: Prentice-Hall, 2002.
[18]	YAN-Chih Huang, “A QoS guarantee streaming system based on SDN/NFV” .  Tamkang University, 2019,  Master’s Thesis.
[19]	A. N. Toosi, R. N. Calheiros, R. K. Thulasiram and R. Buyya, “Resource Provisioning Policies to Increase IaaS Provider's Profit in a Federated Cloud Environment, ” 2011 IEEE International Conference on High Performance Computing and Communications, Banff, AB, 2011, pp. 279-287.
[20]	“Amazon Elastic Compute Cloud (Amazon EC2) ”, [online] Available: http://aws.amazon.com/ec2.